Content uploaded by Christina Beatty
Author content
All content in this area was uploaded by Christina Beatty
Content may be subject to copyright.
1
TWENTY YEARS ON: HAS THE ECONOMY OF THE COALFIELDS
RECOVERED?
Christina Beatty, Stephen Fothergill and Ryan Powell
Centre for Regional Economic and Social Research
Sheffield Hallam University
March 2005
2
Summary
Since the end of the miners’ strike in March 1985, Britain’s coal industry has
experienced an unprecedented loss of jobs. Total employment in the industry has
fallen from over 220,000 to around 7,000. The number of miners has fallen from
170,000 to 4,000. Only 8 of the 170 pits at the time of the strike remain in operation.
This paper examines the extent to which the economy of the coalfields has recovered
from these devastating job losses. It presents new and comprehensive figures on
labour market trends among men in each of the English and Welsh coalfields over
the whole of the period from 1981 to 2004. Five key observations emerge from this
analysis.
First, about 60 per cent of the jobs lost from the coal industry since the early 1980s
have now been replaced by new jobs for men in the same areas. This is strong
evidence of a recovery, and shows that in practice the coalfields have not entered a
‘spiral of decline’. Nevertheless, that still leaves some 90,000 coal jobs still to be
replaced.
Second, the pace of recovery appears to be quickening. The first years of the
present decade have seen several major regeneration initiatives (such as the English
Partnerships National Coalfields Programme) come to fruition at a time of sustained
national economic growth. The rate of new job creation in the coalfields has almost
doubled, with a higher proportion of the new jobs also going to men.
Third, taking a wider view of the employment problems of the coalfields – for instance
to include unemployment inherited from before the pit closures – the extent of
recovery looks a little less impressive: only around half of the overall job shortfall for
men in the coalfields has so far been eliminated. This figure takes account not only
of new job creation but also of the role of out-migration and out-commuting in
stabilising local labour markets.
Fourth, claimant unemployment figures, which are currently relatively low in most
former coalfields, give a wholly misleading view of the strength of the local labour
market. Since the early 1980s, the rise in the number of ‘economically inactive’ men
3
of working age in the coalfields has been twice as large as the fall in recorded
unemployment. In the English and Welsh coalfields in mid-2004, no fewer than
336,000 adults of working age (201,000 men, 135,000 women) were out of work and
claiming incapacity benefits, compared to just 67,000 (50,000 men and 17,000
women) claiming unemployment benefits. The evidence supports the view that in the
coalfields, as in some other parts of older industrial Britain, there has been a huge
diversion of people with health problems from unemployment to incapacity benefits.
Estimates suggest that as many as 100,000 men in the coalfields are currently
‘hidden unemployed’ in this way.
Fifth, there is considerable diversity of experience among the coalfields. Some of the
smaller mining areas in the Midlands, in Leicestershire and Warwickshire for
example, now appear to be well on the way to full recovery. The Yorkshire coalfield
has been an especially encouraging performer in the last few years, though still has
some way to go. In contrast, the large South Wales coalfield in particular has made
slow progress in replacing coal job losses and other areas, such as Northumberland,
also appear to face continuing severe labour market difficulties.
The paper concludes that if the present impressive rate of job creation can be
sustained it will take perhaps another six or seven years to finally replace all the coal
job losses by new jobs for men, and perhaps a further five years to wipe out the
wider accumulated job shortfall. Some areas, like South Wales, look likely to take
rather longer. The progress to date has nevertheless required huge efforts from
local authorities, development agencies, central government and the European
Union, and to finish the job this support will need to be sustained for some while
longer.
4
Background
In March 1985, Britain’s coalminers returned to work after the longest and most bitter
industrial dispute in modern times in the UK. They had lost their battle to stop pit
closures. The year-long miners’ strike was one of the defining events of Margaret
Thatcher’s premiership.
Over the months and years that followed, the pit closures that the miners had feared
did indeed happen. In fact, the closures and job losses were greater than anyone
had predicted. At the time of the strike the state-owned UK coal industry employed
171,000 miners at 170 collieries, and had a total workforce (including white-collar
staff, workshops, opencast mines etc) of 221,000. Nearly 90 per cent of this
workforce was shed during the first ten years after the strike, and job losses have
continued on a smaller scale ever since. The now privatised coal industry employs
fewer than 7,000 in total, of whom only 4,000 work at the eight remaining collieries1.
One of the most important features of this job loss is that it has been virtually all
concentrated in just a dozen or so areas across Britain. This is inevitable given the
nature of mining: coal can only be dug in the places where coal deposits are found.
However, in most of these areas coalmining had been the dominant source of
employment for men, so the consequences for local labour markets were always
going to be serious. In its scale, speed and geographical concentration, the
1 These are the seven remaining deep mines belonging to UK Coal plc – Kellingley, Maltby
and Rossington in Yorkshire, Harworth, Thoresby and Welbeck in Nottinghamshire, and Daw
Mill in Warwickshire - plus Tower Colliery in Wales. In addition there are a number of
extremely small private mines.
5
contraction of the UK coal industry is arguably the definitive example of de-
industrialisation in Britain or Western Europe.
Twenty years on from the end of the strike, to what extent has the economy of the
coalfields recovered? In the intervening years there have been numerous
regeneration initiatives in former mining areas, and the UK economy as a whole has
experienced a sustained period of economic growth since emerging from the
recession of the early 1990s. Has this meant that the economy of the coalfields has
bounced back? Or have these areas remained in the doldrums, by-passed by newer
forms of economic activity? The honest answer is that until now no-one has really
known, at least not in any systematic quantifiable sense. It is this key gap in
knowledge that the present paper sets out to address.
Scope of the study
The present paper focuses on the number of jobs in the coalfields. We make no
apologies for this emphasis. The miners’ strike was about jobs, and it was the loss of
jobs that so marked the subsequent experience of Britain’s mining areas. It is
therefore entirely appropriate to document the extent to which the jobs lost from the
coal industry have been replaced.
More specifically, the paper focuses on jobs held by men. Again we make no
apologies. The coal industry was an overwhelmingly male employer. The loss of
coal industry employment was therefore primarily a problem for men, so it is
appropriate to look at how the male labour market has recovered. At the same time,
it is important to recognise that ‘male’ and ‘female’ labour markets do not operate
entirely independently of each other, so account does needs to be taken of the main
interactions in understanding men’s experience.
The paper covers the 23 years between 1981 and 2004 - a slightly longer period than
the twenty years since the end of the strike. This is a pragmatic. In a previous study,
published nearly a decade ago, two of the present authors examined labour market
adjustment in the UK coalfields over the period 1981-91 (Beatty and Fothergill 1996).
The choice of dates for this earlier study was driven by the availability of fine-grain
data from the decennial Census of Population. What we have done in the present
study is to extend the core elements of this previous work through initially to 2001, to
6
take account of new Census data, and then to 2004 using a range of other data
sources. It is nevertheless worth noting that the three years leading up to the start of
the miners’ strike (ie 1981-84) were also ones of substantial job loss from the coal
industry – some 50,000 coal jobs disappeared during these years. The inclusion of
these earlier years therefore serves to emphasise the scale of the labour market
challenge facing coalfield areas.
Finally, the paper covers only the English and Welsh coalfields. This leaves out the
Scottish coalfields. This is not ideal but has been forced on us by an unusual piece
of negligence. At the time of the original 1996 study, the authors discovered that an
electronic version of the ward-based Special Workplace Statistics for Scotland from
the 1981 Census had ‘gone missing’ from all the main archives, making it extremely
difficult to replicate key parts of the analysis in Scotland. We have not ascertained
whether in the intervening years the relevant electronic data has been ‘rediscovered’.
However, the absence of previous analyses on which to build has meant that for the
time being at least the present paper only covers England and Wales. In population
terms, the English and Welsh coalfields account for about 90 per cent of the UK total.
The central analysis in the present paper, as in the 1996 study, involves the
construction of ‘labour market accounts’ for the coalfields – for the 1981-2004 period
as a whole, for sub-periods, and for each individual coalfield. Labour market
accounting is a tried-and-tested method for disaggregating change in local labour
markets. Its usefulness lies in the fact that changes in employment and
unemployment are not simply and mechanically linked, especially at the local scale.
A fall of, say, 1000 in employment does not lead to a corresponding rise of 1000 in
unemployment. There are several mediating influences – migration, commuting and
changes in labour force participation for example – which mean that in any given
area the changes in employment and unemployment are unlikely to be of the same
magnitude, or even necessarily in the same direction. It is therefore wrong to
assume that because claimant unemployment is now relatively low in most former
mining areas there must have been substantial local job generation. The reality of
local labour market adjustment is much more complex.
The labour market accounts provide an overview of what has happened in the
coalfields. In this paper we then use the information they generate to answer three
key questions:
7
• To what extent have the jobs lost from the coal industry been replaced by
new jobs in the coalfields? This is the most obvious question of all, but what
it ignores is the fact that the coalfields mostly had large-scale unemployment
even before the pit closures began in the 1980s. The job of successful
regeneration was therefore always a lot larger than just replacing coal job
losses. Hence the second question……
• To what extent has the overall job shortfall in the coalfields been eliminated?
This involves taking a wider view of the scale of the coalfield employment
problem. It also involves looking at a wider range of labour market
adjustments, including not only job creation but also out-commuting and out-
migration.
• What is the real level of unemployment in the former coalfields? This involves
looking beyond just the claimant unemployment figures to take account of
people who have been diverted onto other benefits or out of the benefits
system altogether.
Throughout, our focus is on the labour market as a whole in the coalfields, not on the
fate of ex-miners. Many of the older miners who were made redundant in the 1980s
and 90s will now have moved beyond state pension age (65). Other ex-miners will
have found new jobs. The fate of ex-miners is an important research question in its
own right, but the disappearance of mining jobs removed employment opportunities
for later generations as well as for the miners themselves, and when an ex-miner
finds new work he may do so at the expense of other local residents. The loss of
coal jobs was a problem for coalfield communities as a whole, and it is the labour
market as a whole in the coalfields that we investigate.
Defining the coalfields
There is no single definition of ‘the coalfields’. Maps based on geology, historical
connections or labour markets at different points in time would all generate subtly
different definitions.
8
For the purpose of the 1996 study, a ward-based map of the British coalfields was
developed based on the share of resident men in employment who worked in the
coal industry in 1981. The cut-off used was 10 per cent – ie ‘coalfield wards’ were
those that met or exceeded this dependence on coal industry employment, though in
practice there were also a number of minor adjustments to take account of local
circumstances (see Appendix for further details). This definition of the coalfields
subsequently became known as the ‘Sheffield Hallam definition’ and was widely
adopted, for example in government-funded studies of lottery grants (Gore, Dabinett
and Breeze 1999) and educational performance (Gore and Smith 2001). What
should be noted about the Sheffield Hallam definition of the coalfields is that it is
relatively tight, excluding for example areas with only an historic connection with the
coal industry and excluding adjacent urban areas that do not form part of the coalfield
itself. The definition is also firmly rooted in the labour market that existed at the start
of the 1980s, before the pit closures began in earnest. To overcome boundary
changes and allow data to be compared across long periods, the ward-based
definition of the coalfields for subsequent years has been matched as closely as
possible to the original 1981-based definition.
Figure 1 shows the location of the English and Welsh coalfields, based on the
Sheffield Hallam definition. Table 1 shows their current (2001) population. In total,
the English and Welsh coalfields are home to just under 4.5 million people. This
makes them a substantial chunk of Britain, equivalent in population terms to a typical
English region. The Scottish coalfields add a further 0.5 million people, bringing the
total to around 5 million. Three of the coalfields listed in Table 1 (Yorkshire,
Derbyshire and Nottinghamshire) make up a continuous block, extending from just
east of Leeds to near Nottingham, with a population of just under 2 million.
In 1981 the coal industry employed 229,000 men in these English and Welsh
coalfield areas. These jobs accounted for almost exactly one-in-four of all the male
jobs located in these areas. According to 1981 Census figures, these English and
Welsh coalfields already had 160,000 unemployed men even before the main pit
closures began.
9
Figure 1:The English and Welsh coalfields
MILES
100
050
South Wales
North Warwickshire
South Derbyshire/
NW Leicestershire
South
Staffordshire
North
Staffordshire
North
Wales
Lancashire
Derbyshire Nottinghamshire
Yorkshire
Kent
Durham
Northumberland
10
Table 1: The English and Welsh coalfields
Population, 2001
Yorkshire 1,136,000
South Wales 733,000
Durham 535,000
Nottinghamshire 503,000
Lancashire 365,000
Derbyshire 312,000
North Staffordshire 267,000
North Warwickshire 183,000
S Derbys/NW Leics 142,000
Northumberland 140,000
South Staffordshire 109,000
Kent 39,000
North Wales 24,000
England and Wales coalfields 4,488,000
Source: Census of Population
Table 2 shows the loss of coal industry jobs between 1981 and 2004 in each of these
areas. The large Yorkshire coalfield heads the list with the loss of 67,000 male jobs,
followed by Nottinghamshire (40,300) and South Wales (27,200). What is striking in
this table is that in every English and Welsh coalfield well over 90 per cent of all the
coal industry jobs held by men in 1981 disappeared by 2004, and in several areas
coal industry employment was eliminated entirely. In most coalfields, these coal job
losses accounted for a large proportion (typically 20-35 per cent) of all the jobs held
by men in 1981. The exceptions are North Staffordshire and Lancashire, where
coalmining took place in highly urbanised areas alongside other industries and where
as a consequence the coal job losses accounted for a smaller proportion of total
employment.
11
Table 2 : Job loss in coal 1981-2004
Male job loss
1981-2004
as % male coal
jobs in 1981
as % of all male
jobs in area
in 1981
Yorkshire 67,000
95
27
Nottinghamshire 40,300
96
35
South Wales 27,200
97
21
Durham 22,800
100
26
Derbyshire 13,700
99
20
Northumberland 10,100
94
32
S Derbys/NW Leics 9,600
99
31
North Staffordshire 8,600
100
13
Lancashire 7,100
100
9
South Staffordshire 5,700
100
27
North Warwickshire 5,500
92
23
Kent 3,200
100
48
North Wales 1,200
100
39
England and Wales coalfields 222,000
97
24
Sources: Census of Population, Coal Authority
Labour market accounts
Labour market accounts disaggregate local labour market trends into a number of
separate flows:
NATURAL INCREASE IN THE WORKFORCE – the excess in the number of
16 year olds reaching working age over the number of people reaching state
pension age (65 for men, 60 for women) and deaths of working age.
NET MIGRATION – the balance between the number of people of working
age moving into an area and the number moving out.
CHANGE IN NET COMMUTING – the change in the balance of commuting
flows in and out of an area.
12
CHANGE IN LABOUR FORCE PARTICIPATION – the change in the
proportion of adults of working age who are ‘economically active’ (ie in
employment or seeking employment).
CHANGE IN EMPLOYMENT
CHANGE IN RECORDED UNEMPLOYMENT
This approach to local labour markets was first deployed in the UK regional context
by the Cambridge Economic Policy Group (1980, 1982). It was successfully applied
to the UK’s cities by Begg, Moore and Rhodes (1986) and later by Turok and Edge
(1999). In addition to the previous application to the coalfields in the 1996 study, two
of the present authors have also assembled labour market accounts for England’s
disadvantaged rural areas (Beatty and Fothergill 1997) and Britain’s seaside towns
(Beatty and Fothergill 2004a).
One of the advantages of labour market accounts is that all the components are
arithmetically related, so it is possible to show how they work together to generate
the overall pattern of labour market change. There are a number of different ways in
which the accounts can be presented. The one we follow here is as follows:
Job loss in coal
PLUS Natural increase in the workforce
PLUS Net in-migration
PLUS Increase in net in-commuting
PLUS Increase in economically active
MINUS Increase in non-coal jobs
EQUALS Increase in recorded unemployment
In the context of the coalfields, this arrangement of the accounts is particularly helpful
because it exposes how a large loss of jobs from the coal industry has been able to
co-exist with often falling recorded unemployment in the same areas. The first five
lines of the accounts measure the change in labour supply arising from coal job
losses and other labour market trends. The sixth line measures the change in labour
demand arising from sectors other than coal. The change in recorded
unemployment, in the final line of the accounts, is the difference between the two. In
13
this formulation of the accounts, any ‘hidden unemployment’ will be included in the
change in the number of economically active men.
Assembling labour market accounts is a formidable task involving the manipulation of
large amounts of Census and other data. In the analyses presented here, the use of
ward-level data corresponding to the Sheffield Hallam definition of the coalfields is an
added complication. An appendix to the paper describes the detailed data sources
and methods.
Labour market accounts for working-age men, for the English and Welsh coalfields
as a whole between 1981 and 2004, are presented in Table 3. This is a pivotal table
in understanding the process of labour market adjustment in the coalfields and
therefore justifies being explained at some length.
Table 3 : Labour market accounts for working age men, English and Welsh
coalfields 1981-2004
no. as % male
working age
pop. in 1981
Job loss in coal 222,000
15.9
PLUS Natural increase in workforce 86,600
6.2
PLUS Net in-migration -57,900
-4.1
PLUS Increase in net in-commuting -30,700
-2.2
PLUS Increase in economically active -162,500
-11.6
MINUS Increase in non-coal jobs 132,400
9.5
EQUALS Increase in recorded unemployment -74,900
-5.4
Sources : see Appendix
The first point to note is that between 1981 and 2004 the very large loss of coal jobs
(222,000) co-existed with a substantial reduction in recorded unemployment (nearly
75,000). This is not what might have been expected in areas of acute job loss and at
first sight could be taken to indicate exceptionally successful adaptation. The reality
is more complex, as the labour market accounts demonstrate.
Over this twenty-three year period, natural increase in the male workforce in the
coalfields added a further 86,000 to the excess labour supply caused by the loss of
coal jobs. This was partly offset by the loss of 58,000 men of working age through
14
net out-migration – ie. more men of working age moved out of the coalfields than
moved in.
In addition, a net increase in commuting to other areas reduced local labour supply
by another 30,000. The coalfields overall have been substantial net exporters of
commuters to other areas (such as neighbouring cities) for many years – ie, the
number of men and women commuting out exceeds the number commuting in.
Gross flows have tended to increase in both directions as local labour markets have
become less self-contained. A negative ‘increase in net in-commuting’, such as the
one shown in Table 3, represents an increase in the net outward flow.
The really big adjustment in labour supply, and the dominant feature of labour market
change in the coalfields, was however a big reduction in economic activity among
men of working age. ‘Economically inactive’ men are those who are neither in
employment nor recorded as unemployed (in this instance using the Census
definitions of ‘unemployment’). These are the working age men who according to
official figures have dropped out of the labour market altogether, either out of choice
or because circumstances have pushed them out. Between 1981 and 2004 the
reduction in male economic activity rates in the coalfields took no fewer than 162,500
working age men out of the local workforce, equivalent to one-in-nine of the entire
male population between the ages of 16 and 64. This is a large change, whether
viewed in absolute or proportional terms. Later in the paper we take a closer look at
these economically inactive men and argue that in fact a substantial number of them
could be described as ‘hidden unemployed’. For the moment, the point to note is that
over these years the number of men dropping out of the coalfield labour market into
‘economic inactivity’ was more than double the reduction in recorded male
unemployment.
Finally, there was an increase of just over 132,000 in the number of non-coal jobs in
the coalfields held by men. This figure provides firm evidence that a recovery in the
labour market for men in the coalfields is indeed underway.
Table 4 looks at the labour market changes among men in the coalfields over three
sub-periods – 1981 to 1991, 1991 to 2001, and 2001 to 2004. Nearly all the figures
for the first two sub-periods are taken from the Census of Population, which is
comprehensive and mostly reliable. The figures for the 2001-04 period are derived
15
Table 4 : Labour market accounts for working age men, English and Welsh
coalfields by sub-period
1981-91 1991-01
2001-04
Job loss in coal 159,400 60,000
2,500
PLUS Natural increase in workforce 62,100 18,800
5,700
PLUS Net in-migration -59,600 -11,200
13,000
PLUS Increase in net in-commuting -4,500 -22,800
-3,500
PLUS Increase in economically active -84,600 -78,600
700
MINUS Increase in non-coal jobs 44,900 48,300
39,200
EQUALS Increase in recorded
unemployment 27,800 -82,000
-20,700
Sources : see Appendix
from a wider range of sources and involve a greater degree of estimation. The table
reveals at least five important shifts through time.
First, the job loss in coal was front-loaded, with the largest losses in the 1980s and
declining losses thereafter.
Second, migration patterns among working-age men have turned around – from
substantial net out-migration from the coalfields in the 1980s to modest net in-
migration in the latest period. It is tempting to attribute this turn-around to the
declining pace of coal job losses and the growing success of regeneration efforts, but
it is also important to note that net out-migration from the North more generally (and
most coalfield areas are in the North) has ground to a halt in recent years. This owes
more to the growth in international in-migration to the South than to any narrowing in
regional job opportunities, as a recent analysis has shown (Rowthorn 2004).
Third, the withdrawal of men into economic inactivity in the coalfields, which is such a
dominating feature of trends over the whole 1981-2004 period, appears to have
come to a halt. In contrast to the earlier periods, there is no evidence of any
additional withdrawal between 2001 and 2004.
Fourth, the pace of increase in the number of non-coal male jobs appears to be
accelerating. The absolute numbers rather mask this encouraging trend, but in the
16
Table 5: Male labour market accounts for English and Welsh coalfields, 1981-2004
PLUS
PLUS
PLUS
PLUS
MINUS
EQUALS
natural
increase in increase in increase increase in
job loss
increase in net in- net in- economically all non coal recorded
in coal workforce
migration commuting
active jobs
unemployment
Yorkshire 67,000
24,700
-17,600
5,300
-39,300
55,300
-15,200
Nottinghamshire 40,300
11,600
-9,300
-12,100
-18,000
17,400
-4,900
South Wales 27,200
9,000
-9,400
-9,300
-29,900
5,200
-17,600
Durham 22,800
8,000
-5,900
7,100
-20,700
23,900
-12,500
Derbyshire 13,700
1,300
3,300
-3,300
-10,400
7,800
-3,300
Northumberland 10,100
800
500
-6,900
-5,800
-600
-700
S.Derbyshire/N.W.Leicestershire 9,600
2,500
4,600
-1,400
-3,200
12,900
-800
North Staffordshire 8,600
1,900
-8,700
-11,100
-10,200
-13,600
-6,000
Lancashire 7,100
12,600
-15,300
2,200
-15,100
-100
-8,400
South Staffordshire 5,700
3,900
-200
100
-3,400
7,700
-1,600
North Warwickshire 5,500
9,500
-1,400
400
-4,300
12,900
-3,200
Kent 3,200
600
200
-1,000
-1,200
2,100
-300
North Wales 1,200
200
1,400
-700
-1,000
1,500
-400
England and Wales coalfields 222,000
86,600
-57,900
-30,700
-162,500
132,400
-74,900
Sources : see Appendix
17
Table 6: Male labour market accounts for English and Welsh coalfields, as a percentage of 1981 male working age population,
1981-2004
PLUS
PLUS
PLUS
PLUS
MINUS
EQUALS
job loss
natural
increase in increase in increase increase in
in coal increase in net in- net in- economically all non coal recorded
workforce
migration commuting
active jobs
unemployment
Yorkshire 19.0
7.0
-5.0
1.5
-11.1
15.7
-4.3
Nottinghamshire 25.7
7.4
-5.9
-7.7
-11.5
11.1
-3.1
South Wales 11.9
3.9
-4.1
-4.1
-13.1
2.3
-7.7
Durham 13.7
4.8
-3.5
4.3
-12.4
14.3
-7.5
Derbyshire 14.4
1.3
3.4
-3.5
-11.0
8.2
-3.5
Northumberland 23.3
1.8
1.2
-15.9
-13.4
-1.4
-1.6
S.Derbyshire/N.W.Leicestershire 24.7
6.4
11.7
-3.6
-8.3
33.0
-2.0
North Staffordshire 9.4
2.1
-9.6
-12.2
-11.2
-14.9
-6.6
Lancashire 5.9
10.4
-12.6
1.8
-12.5
-0.1
-6.9
South Staffordshire 17.7
12.2
-0.7
0.5
-10.5
24.1
-4.9
North Warwickshire 10.5
18.3
-2.8
0.8
-8.2
24.6
-6.1
Kent 29.5
5.5
1.7
-9.5
-10.8
19.4
-2.9
North Wales 18.7
2.6
22.8
-10.6
-15.9
23.9
-6.3
England and Wales coalfields 15.9
6.2
-4.1
-2.2
-11.6
9.5
-5.4
Sources : see Appendix
18
three years between 2001 and 2004 the number of non-coal jobs held by men in the
coalfields increased by almost as much as in either of the previous two decades.
Fifth and finally, the reduction in recorded male unemployment in the coalfields
occurred entirely in the two later periods. Even so, it is worth noting that the fall in
recorded unemployment among men between 1991 and 2001 (some 82,000) was
almost entirely matched by an additional withdrawal into economic inactivity (more
than 78,000). Only in the final 2001-04 period is the reduction in recorded
unemployment no longer mirrored by rising economic inactivity.
Tables 5 and 6 present labour market accounts for each individual coalfield for the
full 1981-2004 period. These reveal a complex picture that it is not appropriate here
to describe or attempt to explain. Many of the detailed trends reflect specifically local
circumstances, including interactions with neighbouring areas through migration and
commuting patterns. The important point is that the experience of individual
coalfields has differed and this comes through strongly in assessing, below, the
extent to which the coalfields have recovered.
Job replacement
To what extent have the jobs lost from the coal industry been replaced by new jobs in
the coalfields? Earlier we said that this is perhaps the most obvious question of all,
and the simplest if not necessarily the best measure of the extent to which the
economy of the coalfields has recovered.
The labour market accounts, presented earlier in Table 3, provide one answer:
between 1981 and 2004, 222,000 male jobs were lost from the coal industry in these
areas, and this was offset by an increase of 132,400 in male jobs in other industries
and services in the same areas.
Table 7 presents a coalfield by coalfield view. This compares the local job loss in
coal with the increase in non-coal employment in the area, in both cases just for men.
At this scale the answer is clearly a lot more complex. In five coalfield areas all the
jobs lost from the coal industry have been replaced by new jobs for men. Four of
these (S Derbys/NW Leics, S Staffs, N Warwicks and N Wales) are relatively small
coalfields but the fifth (Durham) is one of the larger ones. At the other extreme, in
19
Table 7 : Job replacement in the English and Welsh coalfields, 1981-2004
Male jobs lost
in coal Increase in non-coal
jobs held by men % of coal jobs
replaced
Yorkshire 67,000
55,300
83
Nottinghamshire 40,300
17,400
43
South Wales 27,200
5,200
19
Durham 22,800
23,900
100
Derbyshire 13,700
7,800
57
Northumberland 10,100
-600
0
S Derby/NW Leics 9,600
12,900
100
North Staffordshire 8,600
-13,600
0
Lancashire 7,100
-100
0
South Staffordshire 5,700
7,700
100
North Warwickshire 5,500
12,900
100
Kent 3,200
2,100
66
North Wales 1,200
1,400
100
Sources : Census of Population, Coal Authority, Annual Business Inquiry
three coalfields (Northumberland, N Staffs and Lancashire) none of the jobs lost in
coal had been replaced by 2004. This is because in these three areas net job losses
for men in other industries added to the job losses in coal.
Table 8 looks at the same information from a different angle and shows the number
of coal job losses still remaining to be replaced in 2004. These figures show that in
all just under 90,000 coal jobs remain to be replaced, or 40 per cent of the total lost
from the coal industry between 1981 and 2004. Nottinghamshire and South Wales
together account for half the job losses still to be replaced.
There is therefore evidence, on this particular measure, of real forward progress but
also of a continuing job still to do. However, at this point three digressions are
necessary.
The first concerns the extent to which the new jobs really are adequate replacements
for the jobs lost from the coal industry. One view is that the jobs in the mines were
well-paid whereas the new sources of employment for men – in new factories, call
20
Table 8 : Post-1981 male coal job losses still to be replaced
no
Nottinghamshire 22,900
South Wales 22,000
Yorkshire 11,700
Northumberland 10,100
North Staffordshire 8,600
Lancashire 7,100
Derbyshire 5,900
Kent 1,100
England and Wales coalfields 89,400
Sources : Census of Population, Coal Authority, Annual Business Inquiry
centres, and the service sector – generally compare unfavourably in terms of pay and
conditions. So mere numerical replacement of coal jobs may overstate the true
extent of recovery. It is certainly true that the coal industry paid good wages for
manual labour, though in fairness not always a great deal more than some of the
other heavy industries of the past. On the other hand, structural shifts in the labour
market in the coalfields and elsewhere should not be overlooked. Manual jobs are in
decline but the number of white-collar jobs, including in the public sector, has been
increasing and many of these new jobs do offer better terms and conditions than the
mining jobs of old. The quality of new jobs for men in the coalfields is a legitimate
topic for research, but the presumption should not necessarily be that the new jobs
are so much worse than those they are replacing.
The second digression concerns women’s jobs. So far we have looked at men in
isolation, but that is not how the real-world labour market operates. Although the
coal industry was an almost exclusively male employer, in large segments of the
labour market men compete with women for the same jobs. Table 9 shows the
increase in non-coal jobs in the coalfields, for women as well as men. This shows
that the large increase in jobs held by men has been matched by an equally large
increase in the number held by women. Overall, non-coal employment in the
coalfields grew by 264,000 between 1981 and 2004 – more in fact than the loss of
jobs from the coal industry itself.
21
Table 9 : Increase in non-coal jobs in the English and Welsh coalfields,
1981-2004
1981-91
1991-01
2001-04
Total
1981-2004
Men 44,900
48,300 39,200
132,400
Women 68,500
47,800 15,600
131,800
Total 113,400
96,100 54,800
264,200
annual rate 11,300
9,600 18,300
11,500
% male 40
50 72
50
Sources : Census of Population, Coal Authority, Annual Business Inquiry
What is especially noticeable is that through time the proportion of new jobs going to
men has been increasing – from 40 per cent between 1981 and 1991, to 72 per cent
between 2001 and 2004. At first sight this is surprising because the industries and
occupations that have shown the strongest employment growth in the UK in recent
years have tended to be those where women are especially concentrated. What the
figures on male and female employment suggest is that the coalfield labour market
may be operating in new ways: whereas ex-miners themselves shunned employment
they saw as ‘women’s work’, the generation of men behind them has adopted a more
open-minded attitude and begun to fill jobs that once would have been taken by
women. If this interpretation is correct it has important implications. First, one of the
very long-term effects of coal industry job losses is to squeeze the job opportunities
for women in the coalfields, as men begin to compete for and take the same jobs.
Second, the increase in the number of jobs for women in the coalfields is opening up
job opportunities for men as well.
The final digression concerns the relationship between output and employment.
There are no figures available to chart the value of economic output (measured for
example by gross value added) at the local scale in the coalfields or elsewhere. The
fall in UK coal production – down around 70m tonnes a year since the early 1980s –
will by itself have reduced the output of the coalfield economy by between £2.5bn a
year (at current coal prices) and £4bn a year (at the prices prevailing in the 1980s).
On the other hand, output per head in virtually all sectors of the economy, nationally
and locally, has been increasing through time. It is possible to be confident therefore
that with male employment in the coalfields clawing its way back towards previous
22
levels and a big increase in female employment, the total value of output from the
coalfield economy will now significantly exceed pre-pit closure levels. In this limited
sense, at least, the economy of the coalfields has recovered. But this does not
detract from the fact that the key coalfield problem has for at least twenty years been
about jobs, and it is in terms of jobs that the true extent of recovery therefore needs
to be judged.
The overall job shortfall
To what extent has the overall job shortfall in the coalfields been eliminated? We
noted earlier that mere replacement of the jobs lost in the coal industry is not a very
good measure of coalfield regeneration. The problem is that the coalfields began to
lose coal jobs when they were already affected by large-scale unemployment. Some
of this pre-existing unemployment can be traced back to coal job losses in earlier
years, but it also reflects the state of the national economy. The year our empirical
investigation starts – 1981 - falls in the middle of the deepest recession to hit the UK
economy in the second half of the twentieth century. The recession was especially
severe in Britain’s manufacturing heartlands, including in coalfield areas where major
job losses often occurred in the factories brought in by previous rounds of regional
policy.
A wider view of the employment shortfall for men in the coalfields involves adding
together:
• The job loss in the coal industry
• The pre-existing unemployment in the coalfields
• The natural increase in the local workforce
This last variable – natural increase – is relevant because it measures the extra jobs
required just to keep pace with the growth of the local workforce. Offsetting the job
shortfall are three factors:
• The increase in non-coal jobs
• Net out-migration
• The increase in net out-commuting
23
The second and third of these variables are relevant because they represent
legitimate ways in ways local labour markets adjust to change, for example as places
adopt new roles as commuter settlements or shrink in population to reflect a smaller
economic base.
Table 10 works through this calculation for men in the English and Welsh coalfields
as a whole over the 1981 to 2004 period. This table shows that the overall job
shortfall for men in the coalfields (nearly 470,000) was more than double the loss of
coal jobs, not least because the coalfields began the period with 160,000
unemployed men. The table also shows that even after the substantial labour market
adjustments of the last two decades, more than half the shortfall (nearly 250,000) still
remains. The growth in non-coal jobs for men has been the largest factor in the
reduction in the shortfall, but out-migration and to a lesser extent out-commuting
have contributed as well.
Table 10 : Job shortfall among men in the English and Welsh coalfields,
1981-2004
no
Job loss in coal 222,000
PLUS 1981 unemployment 160,000
PLUS Natural increase in male workforce 86,600
EQUALS TOTAL JOB SHORTFALL 468,600
MINUS Increase in non-coal jobs 132,400
MINUS Net out-migration 57,800
MINUS Increase in net out-commuting 30,800
EQUALS REMAINING JOB SHORTFALL 247,600
Sources : see Appendix
Table 11 shows the results of replicating this calculation for each individual coalfield.
The share of the shortfall eliminated by 2004, in the second column of the numbers,
arguably provides the single best measure of the extent to which the labour market of
each area has ‘bounced back’. It reveals a diversity of experience, though one that is
consistent with casual observation of the economic health of most of the areas. Two
small Midlands coalfields (S Derbys/N W Leics and N Warwicks) show the strongest
24
Table 11 : Job shortfall among men, by coalfield 1981-2004
Total job shortfall
1981-2004
% eliminated by
2004
Yorkshire 127,800
53
South Wales 68,400
35
Nottinghamshire 65,000
60
Durham 56,400
40
Lancashire 35,200
37
Derbyshire 23,900
33
North Staffordshire 21,300
29
North Warwickshire 21,100
66
Northumberland 15,400
38
S Derbys/NE Leics 14,500
67
South Staffordshire 13,000
60
Kent 4,700
62
North Wales 2,100
32
England and Wales coalfields 468,600
47
Source: Authors’ estimates
recovery. We have already noted that in these two areas all the coal job losses have
been replaced. At the other end of the spectrum, the large South Wales coalfield is a
weak performer on this indicator, as are the Northumberland, Durham and
Derbyshire areas and the more urban coalfields of Lancashire and North
Staffordshire. The continuing labour market weakness in Durham, where we
previously noted that coal job losses have been entirely replaced by new male jobs,
is a reflection of other powerful influences on the local labour market (inherited
unemployment, commuting patterns etc). Durham illustrates the more general point
that mere replacement of coal jobs is not necessarily an indicator of complete
recovery.
In practice it is unrealistic to expect the whole of any job shortfall to be eliminated,
even in the most favourable circumstances. This is because even in the most
prosperous areas there is always residual unemployment as people move between
jobs and as a result of skill mismatches. In addition, economic inactivity among
25
working age men has tended to increase through time irrespective of any surge in
hidden unemployment, for example because of voluntary early retirement and
because of a long-term trend for the numbers on incapacity benefits to rise even in
the most prosperous areas. These labour supply factors impose constraints on the
extent to which the job shortfalls identified in our calculations could ever really be
removed. In that sense, some of the ‘missing’ jobs should not really be counted as
part of the job shortfall at all. It is impossible to be precise, but if the irreducible job
shortfall accounted for 5 per cent of working age men in the coalfields, that would
point to a figure of 70-80,000, or about one-third of the remaining shortfall for men
identified in the calculations. Looked at from the other direction, that means an extra
175,000 male jobs are still needed in the English and Welsh coalfields.
Unemployment
What is the real level of unemployment in the coalfields? One of the striking features
at the present time is the relatively low level of claimant unemployment. Figure 2
shows claimant unemployment among men in the English and Welsh coalfields from
1984 (when ward-level figures first became available) until 2004. The key feature is
the large reduction, initially in the second half of the 1980s and then again from
around 1993. Overall, claimant unemployment among men in the coalfields has
fallen by three-quarters from peak levels. The numbers in the coalfields have broadly
followed the trend in the UK as a whole, though the rate in the coalfields has always
remained a little above the corresponding national figure.
The problem with claimant unemployment data is that it counts only those people
who are out-of-work and claiming unemployment-related benefits, principally
Jobseeker’s Allowance. What the figures fail to include are the unemployed people
who have been diverted onto other benefits or out of the benefits system altogether,
many of whom are conventionally counted as ‘economically inactive’.
Table 12 disaggregates the economic inactivity among men in the coalfields recorded
by three successive Censuses of Population. Earlier, we noted that the increase in
the number of inactive men represented the single largest labour market adjustment
in the coalfields. Table 12 shows that between 1981 and 2001 these men doubled
as a proportion of the total male population of working age, to more than one-in-five
of all 16-64 year olds. A surge in ‘permanent sickness’, to nearly one-in-nine of all
26
Figure 2: Male claimant unemployment in the English and Welsh coalfields,
1984-2004
0
50,000
100,000
150,000
200,000
250,000
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Figures refer to April each year
Source: NOMIS
Table 12 : Economic inactivity among men in the English and Welsh
coalfields
as % of males aged 16-64
1981
1991
2001
Permanently sick 4.9
9.0
10.8
Retired 1.5
3.7
3.1
Students 3.8
3.6
3.7
Looking after family/home )
1.3
) 0.5
0.9
Other inactive )
3.0
All inactive 10.7
17.2
21.9
Source : Census of Population
27
working age men, represented the largest component of the increase. The number
of retired – in this instance ‘early retired’ because all are below state pension age –
also increased, though in the 1980s rather than the 1990s. The number of
economically inactive students remained broadly stable. The number of ‘other
inactive’ increased sharply between 1991 and 2001, though this is probably
attributable mainly to the use of the ‘ILO’ unemployment criteria for the first time in
the 2001 Census. This had the effect of classifying some people who would have
been recorded as ‘unemployed’ in earlier years as ‘inactive’ in 2001 because they
failed to meet one or more of the ILO criteria.
In April 2004, claimant unemployment in the English and Welsh coalfields stood at
67,000 (made up of 50,000 men and 17,000 women). Coalfield unemployment on
the ILO measure (which counts everyone who is out-of-work, available to start in two
weeks, and has looked for work in the last four weeks) was somewhat larger – an
estimated 93,000, comprising 58,000 men and 35,000 women. The ILO figures are
in theory the government’s preferred measure of unemployment. More importantly,
official figures show that in August 2004 no fewer than 336,000 coalfield residents of
working age (200,000 men and 136,000 women) were out-of-work and claiming
incapacity benefits, mainly Incapacity Benefit itself. Or to put the figures another
way, there are five times as many people of working age in the coalfields who are
out-of-work and claiming incapacity benefits as there are out-of-work and claiming
unemployment benefits. The people claiming unemployment and incapacity benefits
are two mutually exclusive groups – it is not possible to claim both benefits at the
same time. Most incapacity claimants do not actively look for work, so they also tend
to be excluded from the ILO unemployment count.
Table 13 shows the distribution of male incapacity benefit claimants by coalfield.
What is noticeable here is that in all but three small Midlands coalfields, these
claimants account for 10 per cent or more of all 16-64 year old men. In South Wales,
the coalfield with the highest density of incapacity claimants, the proportion reaches
nearly one-in-five of all working age men.
The key question concerns the extent to which the very large number of economically
inactive men, and of incapacity claimants in particular, hides unemployment. The
1996 study of the coalfields, mentioned earlier, argued that a substantial proportion
of the men in the coalfields who are recorded as permanently sick should indeed be
regarded as ‘hidden unemployed’. The logic here was not that the benefit claims
28
were fraudulent or the health problems anything less than real, but that in a genuinely
fully-employed economy many of these men would have been in work. This claim
has subsequently been the basis of substantial research in the coalfields and
elsewhere (see for example Alcock et al 2003). The government itself accepts that,
in the right circumstances, many incapacity claimants could hold down jobs.
Table 13 : Male incapacity benefit claimants, August 2004
no. as % of working age
men
South Wales 44,400 19.3
Durham 27,700 16.2
Lancashire 17,500 14.7
North Staffordshire 12,100 14.1
Northumberland 6,300 14.1
North Wales 1,000 13.0
Yorkshire 46,400 12.8
Derbyshire 12,700 12.7
Nottinghamshire 18,600 11.6
Kent 1,200 10.2
North Warwickshire 6,000 9.9
South Staffordshire 3,300 9.1
S Derbys/NW Leics 3,500 7.5
England and Wales coalfields 200,600 14.0
Sources :DWP and ONS
Since the publication of the 1996 coalfield study, improved methods have been
developed to estimate the numbers of incapacity claimants who ‘could reasonably be
expected to have been in work in a fully-employed economy’. These methods are
based on establishing a ‘benchmark’ for every area that combines:
• The incapacity claimant rates currently prevailing in the parts of southern
England at or close to full employment (a group of seven counties to the
north, west and south of London)
29
• The underlying differences in rates of incapacitating ill health between each
area and this fully employed part of southern England (using the differences
recorded by the 1981 Census, before the figures became badly contaminated
by the diversion from unemployment)
In each area, incapacity claimant numbers in excess of this benchmark are judged to
represent hidden unemployment.
These methods are explained in full elsewhere (see in particular Beatty and Fothergill
2004b) and the robustness of the resulting estimates has been tested using
alternative methods. What we have done here is to apply these methods to the
coalfields, using ward-level incapacity claimant data for August 2004 from the
Department for Work and Pensions. For each individual coalfield this has involved
establishing a benchmark that reflects claimant rates in fully employed parts of Britain
and the underlying higher rate of incapacitating ill health in the local area.
Take the example of Yorkshire. In this coalfield, a total of 46,400 men were out-of-
work and claiming incapacity benefits in August 2004 – 12.8 per cent of the entire
male working age population. The benchmark for this area is 6.5 per cent - the sum
of the present-day incapacity claimant rate among men in fully-employed parts of the
South (4 per cent) and the difference in the rate of ‘permanent sickness’ between the
Yorkshire coalfield and this part of the South recorded in 1981 (2.5 per cent). The
difference between the actual claimant rate and the benchmark – 6.3 per cent,
equivalent to 23,000 men – is estimated to be a form of hidden unemployment. This
calculation takes full account not only of what has already been shown to be possible
in the parts of Britain where the local economy is very strong, but also of the fact that
incapacitating ill health is more widespread in the Yorkshire coalfield than in the fully-
employed parts of the South.
The results of this exercise are presented for the coalfields as a whole in Table 14.
This shows that just over 100,000 men in the English and Welsh coalfields are
estimated to have been diverted from unemployment onto incapacity benefits. These
men represent almost exactly half of the 200,000 men of working age in the
coalfields claiming incapacity benefits. The number estimated to be diverted from
unemployment to incapacity benefits is double the number of male claimant
unemployed in the same areas. Adding this group and the extra ILO unemployed to
30
Table 14 : Unemployment among men in the English and Welsh
coalfields, mid-2004
no. as % male working
age population
Claimant unemployed 50,300 3.5
Additional ILO unemployed 7,500 0.5
Diverted to incapacity benefits 101,400 7.1
‘Real’ unemployment 159,200 11.1
Sources : ONS and authors’ estimates
claimant unemployment more than triples estimated total unemployment among
men2.
Table 15 presents the results of the same calculation for each individual coalfield. In
all areas, the ‘real’ level of unemployment is markedly higher than claimant
unemployment, reflecting the scale of hidden unemployment especially among
incapacity claimants. The inclusion of hidden unemployment also differentiates
individual coalfields more sharply. Rather than being tightly clustered with claimant
unemployment rates between two and five per cent, the inclusion of hidden
unemployment widens the range to 5-13 per cent. South Wales and the two North
East coalfields appear especially disadvantaged on this wider measure of
unemployment.
Too much reliance should not be placed on the precise figures. The methodology
used to estimate hidden unemployment among incapacity claimants is based on a
number of assumptions and in practice there will have been changes in the
underlying level of incapacitating ill health in different areas. The figures we present
are intended to provide broad estimates of the scale of the distortion to conventional
unemployment data. Nevertheless, this alternative perspective on unemployment
casts the coalfields in a very different light. It suggests that rather than having been
2 The definition of ‘real’ unemployment used in Table 14 (and in Table 15) is slightly narrower
than the one used in the 1996 coalfield study and in other published estimates by Beatty-
Fothergill team in that it excludes diversions from claimant unemployment into ‘early
retirement’ and onto ‘government schemes’. Neither of these additional diversions is now
likely to be large given the falling numbers in both groups and the changing nature of most
government schemes. However, because of this methodological change the figures
presented here should not be compared with earlier estimates of real unemployment for the
coalfields or other areas.
31
Table 15 : Unemployment among men by coalfield, mid-2004
Claimant unemployment ‘Real’ unemployment
no. % working
age no. % working
age
South Wales 8,800
3.8
31,100
13.5
Durham 7,500
4.4
22,700
13.3
Northumberland 2,200
4.9
5,900
13.2
Lancashire 4,600
3.8
14,100
11.9
N Staffordshire 2,900
3.4
9,900
11.6
Derbyshire 3,700
3.7
10,900
10.9
Yorkshire 11,900
3.3
36,500
10.1
Nottinghamshire 5,000
3.1
15,800
9.9
North Wales 200
2.0
700
9.1
N Warwickshire 1,500
2.4
5,400
8.9
S Staffordshire 1,000
2.7
2,700
7.5
Kent 300
3.0
800
7.2
S Derbys/NW Leics 700
1.6
2,500
5.4
England and Wales coalfields 50,300
3.5
159,200
11.1
Sources : ONS and authors’ estimates
reduced to modest levels, unemployment among men in the coalfields has mostly
become hidden from view, above all diverted onto incapacity benefits.
This radically different perspective actually fits well with official data on ‘employment
rates’ – that is, the share of adults of working age who have jobs. The 2001 Census
recorded an employment rate among men for Britain as a whole of 74 per cent, and
in the South East of England of 80 per cent. By contrast, the average for the
coalfields was just 70 per cent. No coalfield was able to match the South East
employment rate and only three small Midlands coalfields (S Derbys/N W Leics, N
Warwicks and S Staffs) exceeded the national rate. The South Wales coalfield – the
worst on our wider measure of unemployment – had a male employment rate of just
64 per cent, ten percentage points behind the national rate and sixteen percentage
points behind the South East of England. Differences in male employment rates
between areas primarily reflect differences in economic activity rates, which in turn
primarily reflect differences in incapacity claimant rates. Our view is that while health
factors do play a part, local labour market conditions are the key to understanding the
32
scale of the disparities. South Wales has an employment rate so far adrift of the
South East of England, and an incapacity claimant rate so much higher, primarily
because the demand for labour in South Wales is so much weaker.
What we are arguing is that one of the main long-run adjustments to coal job losses
has been the diversion onto incapacity benefits. This constitutes a form of hidden
unemployment because many of the men claiming these benefits would have been in
work in a fully-employed economy. Indeed, prior to the pit closures far fewer men in
the coalfield claimed these benefits. Underlying health problems have not radically
worsened (indeed, national trends show some gradual improvement), which
suggests that at one time far more people with health problems did indeed hold down
jobs. But many of these jobs, especially in the coal industry of course, have
disappeared. In the competition for the remaining jobs, people with health problems
are amongst those who have lost out and the differences in benefit payment rates
has created an incentive for them to claim Incapacity Benefit (to which they are
entitled on health grounds) rather than Jobseeker’s Allowance.
At the same time it is important to acknowledge that this form of hidden
unemployment differs in several respects from conventional unemployment. In
particular, once on Incapacity Benefit most men and women quickly give up looking
for work. They are not therefore part of the pool of active jobseekers from which
employers fill vacancies. This has implications for policies to re-engage these
individuals with the labour market because although shortfalls in the local demand for
labour act as the trigger for rising incapacity benefit numbers, subsequent job
creation does not automatically bring the numbers down again. It is the active
jobseekers, such as those claiming unemployment benefits, who take up the job
opportunities. This is precisely what has happened in the coalfields in recent years:
as new jobs have been created, claimant unemployment has fallen but the numbers
on incapacity benefits have hardly been dented. To make in-roads in future to the
stock of incapacity claimants will require job creation to go hand-in-hand with labour
market activation policies. Forcing incapacity claimants to look for work without at
the same time ensuring an adequate supply of suitable jobs in the right places is
unlikely to succeed. But equally, further job creation without practical support for
incapacity claimants to re-enter employment is unlikely to work either.
33
An assessment
Let us now try to provide an answer to the paper’s central question: has the economy
of the coalfields recovered?
The first and most fundamental point that needs to be made is that the English and
Welsh coalfields have not entered a ‘spiral of decline’. On the contrary, there is
incontrovertible evidence that the labour market in the coalfields is bouncing back
from the hammer-blow of coal job losses. This is encouraging, especially as there
was never any certainty that this would be the trajectory.
The more difficult issue is the scale of the recovery. The answer on this point varies
according to exactly which yardstick is used. In terms of the number of coal jobs
replaced by new jobs for men, the coalfields as a whole appear to be about sixty per
cent along the way to full recovery. On the other hand, taking a wider view of the
employment shortfall (including for example the high inherited level of
unemployment) suggests that only about half the job is done. Low claimant
unemployment, however, presents a wholly misleading impression of the extent of
recovery. In practice there has been a huge diversion of working age men into
‘economic inactivity’, and in particular onto incapacity benefits. Much of this diversion
should be regarded as ‘hidden unemployment’.
Looking at these findings as a whole suggests that, as a sweeping generalisation, the
economy of the coalfields is perhaps a little over half way towards full recovery. The
coalfields have come a long way, but still have a long way to go. An additional
90,000 male jobs would be needed to claw back all the job losses in the coal industry
since the early 1980s, and another 50,000 or more to erode the inherited male job
shortfall.
The extent to which recovery can be attributed to national economic growth or to
interventions to help the coalfields is unclear. In truth, both have probably
contributed in major ways, but it would need a further major piece of research to
unravel the respective contributions. What is clear is that since the beginning of the
present decade the combination of a long period of national growth and the coming-
to-fruition of key regeneration measures conceived in the 1990s (the English
Partnerships coalfield programme and Enterprise Zone status for example) has
accelerated the pace of recovery.
34
What is also clear is that the experience of individual coalfields has varied a great
deal. The contrast between Yorkshire and South Wales is the most striking example.
These are the two largest coalfields in terms of population, with 1.1 and 0.7 million
people respectively. The whole of the South Wales coalfield and the largest part of
the Yorkshire coalfield both presently enjoy EU Objective 1 status and its equivalent
Tier 1 status for regional aid from the UK government. Yet here the similarities
appear to end. Whereas in Yorkshire there has been a net increase of 55,000 in the
number of non-coal jobs held by men since 1981, in South Wales the increase has
been only 5,000. The performance of Yorkshire has been particularly impressive in
recent years, with non-coal jobs for men up by 30,000 between 2001 and 2004
alone.
The statistical disparity between Yorkshire and South Wales chimes with casual
observation. The Yorkshire coalfield, bounded by motorways on all sides and with
new road links into its core, is reasonably centrally located in Britain as a whole and
was perhaps always a candidate for recovery once the basic building blocks (roads,
sites, training etc) had been put in place. There has been major new investment in
the Dearne Valley Enterprise Zone in South Yorkshire, along the M62 in West
Yorkshire, and on the outskirts of Doncaster, to mention just three examples. Much
of this has happened since the late 1990s after substantial preparatory investment.
The South Wales coalfield, by contrast, appears to be much more intractable case.
Whilst South Wales has a longer history of regeneration initiatives, the difficult terrain
of the mining valleys and the greater isolation of many of communities seem never to
have proved very attractive to new investment. In so far as development taken place
in South Wales in recent years this has often been along the M4 corridor, just off the
coalfield itself. As a result, the trajectory of the South Wales valleys has been less
one of job replacement and more one of rising out-commuting. In the valleys,
employment rates remain low and the real level of unemployment very high.
The statistics for the Durham coalfield tell a different story. In this part of the country
new job creation for men has been impressive, to the extent that on our figures all the
coal jobs lost since the beginning of the 1980s have been replaced. A new Nissan
car factory, and major call centre developments at Doxford Park on the outskirts of
Sunderland and in Easington district have been among the flagship investments. But
the Durham coalfield is embedded in a region with more widespread employment
problems, so as jobs have been created in the former coalfield there has been rising
35
in-commuting from neighbouring areas. This has reduced the benefit to coalfield
residents. One consequence is that a large imbalance in the Durham coalfield labour
market still persists, with much of the imbalance hidden away in the very large
numbers on incapacity benefits.
The Northumberland coalfield shares the Durham coalfield’s continuing and serious
labour market imbalance but the dynamics of its local labour market are different. In
Northumberland there has been little success in increasing the stock of jobs for men.
The consequence has been rising out-commuting, particularly to the neighbouring
Tyneside conurbation, and it is this that has kept an already weak labour market from
becoming much worse.
The clear success stories are the three small coalfields of the Midlands – South
Derbyshire/North West Leicestershire, North Warwickshire, and South Staffordshire.
Here the coal job losses have been fully replaced and good progress has been made
towards eliminating overall job shortfalls. Local unemployment is now relatively low,
even taking a wider view. However, the small scale and central location of these
coalfields, and their proximity to areas of prosperity, always meant that regeneration
in these areas was never going to be as daunting a task as elsewhere.
So how much longer can full recovery in the coalfields be expected to take? The
recent rate of job creation has been impressive – more than 50,000 new jobs in the
coalfields between 2001 and 2004, of which more than two-thirds went to men. If this
rate of job creation is sustained it will take only another six or seven years to replace
all the coal job losses by new jobs for men, and perhaps a further five years to wipe
out the wider accumulated job shortfall. With claimant unemployment relatively low
in most areas, much of the discernible impact from here on would take the form of
lower numbers of economically inactive men, and lower numbers on incapacity
benefits in particular. However, individual coalfields where difficulties are
entrenched, such as South Wales, could be expected to take rather longer to make a
full recovery even with the right measures in place.
Whether recent rates of job creation can be sustained depends a great deal on
continuing growth in the national economy. The stability of UK economic growth
since the mid 1990s has been astonishing but of course there can be no guarantee
that the future will mirror the recent past. The UK’s large trade deficit and high levels
of household debt are worrying signs. But even if national economic growth were to
36
be sustained it would be foolish to rely on market forces alone to resolve the
coalfields’ remaining economic problems. The coalfields have come this far along
the road to recovery in part because of the intensive practical support they have
received from local authorities, development agencies, central government and the
European Union. To finish the job, this support will need to be sustained a little while
longer.
37
References
Alcock, P., Beatty, C., Fothergill, S., Macmillan, R. and Yeandle, S. (2003) Work to
Welfare: how men become detached from the labour market, CUP, Cambridge.
Beatty, C. and Fothergill, S. (1996) ‘Labour market adjustment in areas of chronic
industrial decline: the case of the UK coalfields’, Regional Studies, vol 30, pp 627-40.
Beatty, C. and Fothergill, S. (1997) Unemployment and the Labour Market in Rural
Development Areas, Rural Research Series no. 30, Rural Development Commission,
London.
Beatty, C. and Fothergill, S. (2004a) ‘Economic change and the labour market in
Britain’s seaside towns’, Regional Studies, vol 38, pp 461-480.
Beatty, C. and Fothergill, S. (2004b) The Diversion from ‘Unemployment’ to
‘Sickness’ Across British Regions and Districts, CRESR, Sheffield Hallam University.
Begg, I., Moore, B.C. and Rhodes, J. (1986) ‘Economic and social change in urban
Britain and the inner cities’, in Hausner V. (ed) Critical Issues in Urban Economic
Development, Clarendon Press, Oxford.
Cambridge Economic Policy Group (1980) ‘Urban and regional policy with provisional
regional accounts for 1996-78’, Cambridge Economic Policy Review , no 6.
Cambridge Economic Policy Group (1982) ‘Employment problems in the cities and
regions of the UK: prospects for the 1980s’, Cambridge Economic Policy Review, no
8.
Gore, T., Dabinett, G. and Breeze, J. (1999) Coalfields and the Lottery, DCMS,
London.
Gore, T. and Smith, N. (2001) Patterns of Educational Achievement in the British
Coalfields, research report 314, DfES, London.
Rowthorn, R. (2004) Combined and Uneven Development: reflections on the North-
South divide, paper presented to the HMT/DWP annual labour market issues
seminar, London, October 2004.
Turok, I. and Edge, N. (1999) The Jobs Gap in Britain’s Cities, Policy Press, Bristol.
38
Appendix: data sources and methods
The Sheffield Hallam definition of the coalfields
The coalfields are defined here as those contiguous groups of wards where at least
10 per cent of the resident males in employment in 1981 were engaged in the energy
and water sector (which in these areas overwhelmingly comprised coalmining). In
practice this definition has been adjusted to include some wards that did not meet the
10 per cent threshold but which were largely or wholly surrounded by other coalfield
wards. Also, in the Lancashire and North Staffordshire areas, where mining took
place in a more urban context, some wards with slightly lower dependence on coal
are included. Two isolated wards (in West Cumbria and Durham) meeting the 10 per
cent criteria but separated from the main coalfields are excluded.
This definition is based on the coalfield labour market at a particular point in time,
which is appropriate for a labour market study. It covers virtually all the mines
working in 1981. It does however exclude a number of areas with weaker or more
historic connections with the coal industry (for example in West Durham, West
Cumbria, Lancashire and the Forest of Dean) that are legitimately included in other
coalfield definitions.
Ward boundaries in 1991 (for 1991 Census data) and 2003 (for 2001 Census and
other recent data) have been matched as closely as possible to the 1981 ward-based
definition of each coalfield. This facilitates the use of data for different points in time
on different ward boundaries.
Labour market accounts
The data for the labour market accounts for 1981-91 and 1991-2001 come
overwhelmingly from the Census of Population. The data for the more recent 2001-
04 period comes from a variety of sources and is inherently less reliable or
comprehensive than Census data, for example because some elements are based
on sample surveys such as the Labour Force Survey (LFS) and the Annual Business
Inquiry (ABI). Additionally, some of the post-2001 data (eg from the LFS) is only
available for districts or larger units. Figures for 2001 to 2004 are consequently
subject to a greater margin of error than the figures for earlier years. The resulting
distortion to the figures for the whole 1981-2004 period is however likely to be
modest.
The labour market accounts for 1981-2004 are the summation of separate accounts
for each of the three sub-periods (ie 1981-91, 1991-2001 and 2001-04). The detailed
data sources and methods for each element of the labour market accounts are set
out below.
39
JOB LOSS IN COAL
The figures for coal industry employment in 1981 and 1991 are taken from the
Census of Population and refer to total employment in SIC 1.1. The figures for 2001
and 2004 are taken from Coal Authority data on coal industry employment by area.
In all cases the figure refer to jobs actually located in the area, not employees living
in the area.
NATURAL INCREASE IN THE WORKFORCE
This is the excess number of persons reaching working age (ie 16) over the number
leaving the workforce through death or attainment of state pension age (65 for men,
60 for women).
For 1981-91 and1991-2001 natural increase is calculated as the difference between
the actual resident population of working age at the end of the period and the
projected resident population of working age derived from a cohort survival model.
Ten-year, district-level survival rates have been calculated for men using ONS Vital
Statistics for local authority districts and the base-year resident population for each
district. The appropriate district survival rates are applied to the male age structure
at the start of each period (from the Census) of the coalfield wards in each district.
For the period after 2001, the same rate of natural increase as in 1991-2001 has
been rolled forward to 2004. A comparison of population structures in the coalfields
in 1991 and 2001, and the relative absence of volatility in trends in natural increase,
suggest this assumption is acceptably robust.
NET IN-MIGRATION
This is the difference between the actual population of working age and the projected
population of working age derived from the cohort survival model.
For 1991 and 2001 the actual population of working age in each coalfield is the sum
of ward-level data from the Census of Population. An important complication arises
from the treatment of students studying away from home. In the 1991 Census they
were recorded at their home address; in 2001 they were recorded at their term-time
address. To avoid distorting population change between the two dates, students
studying away in 2001 have been added back into the population of their home area.
Given the available Census data it is not possible to also remove students from the
population of their term-time areas. However, the English coalfields have virtually no
higher education establishments likely to give rise to distortions. The important
exception is Keele University, which falls within a coalfield ward in North
Staffordshire, and the figures for this area should therefore be treated with more
caution than those for other areas. The effect of the inclusion of students at their
term-time address is to boost estimated in-migration, to lower observed economic
activity rates, and to raise the number of jobs recorded in the area in so far as
students engage in part-time employment.
For 2001-04 the change in population used to estimate migration is based on the
trend change in working age population between 2001 and 2003, shown by ONS
mid-year population estimates, projected forward to 2004. This exercise uses the
weighted average change in 34 ‘principal coalfield districts’. These include districts
where most or all of the wards are classified as ‘coalfield wards’ (eg Barnsley,
40
Bolsover) but exclude those where coalfield wards comprise only a small part of a
much larger district (eg Leeds).
INCREASE IN NET IN-COMMUTING
This is the difference between net in-commuting at the beginning and end of each
period.
In practice, the coalfields have been substantial net exporters of commuters for many
years so it is worth underlining that this part of the accounts only measures change
through time. A negative increase in net in-commuting represents an increase in the
net commuting outflow; a positive figure represents a reduction in the net outflow.
Net in-commuting is calculated by subtracting the number of residents in employment
from the number of employed in the area (see below). For 1981, 1991 and 2001 the
calculation is based on ward data from the Census of Population.
For 2001-04 the change in net commuting is calculated as the residual in the
accounts.
INCREASE IN ECONOMICALLY ACTIVE
This is the addition to, or reduction in, labour supply resulting from change in the
economic activity rate among persons of working age.
‘Economic activity’ includes the employed, self-employed, recorded unemployed and
temporarily sick. For 1981, 1991 and 2001 the economic activity rate is taken from
ward-level Census of Population data. The change in the economic activity rate is
multiplied by the male population of working age at the end of each sub-period to
provide an absolute number.
The change in economic activity rates between 2001 and 2004 is the weighted
average change in the 34 principal coalfield districts, in each case based on four
quarters’ data from the LFS. The data for individual districts from the LFS tends to
be unreliable because of sampling errors, but pooling observations for a number of
districts and quarters gives a more reliable picture for the coalfields as a whole.
INCREASE IN EMPLOYMENT
This is the change in the number of jobs (including self-employed) located in coalfield
wards.
The source for 1981, 1991 and 2001 is ward-level data from the Census of
Population. Those with ‘no fixed workplace’ and ‘workplace inadequately described’
are included at their place of residence in each year.
The percentage change in employment between 2001 and 2004 is taken from ward-
level data from the ABI (actually from the annual ABI surveys in December 2000 and
December 2003). ABI employment levels differ in detail from those recorded by the
Census because of the different treatment of the self-employed, double-jobbing,
workers above and below normal working age and sampling error. For the coalfields
as a whole, however, and for the larger individual coalfields, the ABI percentage
41
change nevertheless offers a reliable guide. The percentage change has been
applied to the 2001 Census employment figure to generate an absolute number. A
small additional adjustment has also been made to the figures for the Durham
coalfield to take account of detailed boundary changes.
INCREASE IN RECORDED UNEMPLOYMENT
This is the change in the number of persons of working age recorded as
unemployed.
For 1981, 1991 and 2001 the figures are taken from ward-level data in the Census of
Population. The 1981 and 1991 Censuses counted everyone who was out-of-work
and looking for work as ‘unemployed’. The 2001 Census used the ILO measure of
unemployment, counting those who were not only actively looking for work but also
available for work. This difference in methods slightly reduces recorded
unemployment in 2001 relative to earlier years. Both methods record substantially
higher numbers of unemployed than the claimant count, which is based on benefit
numbers, but exclude the jobless who have been diverted into economic inactivity.
In 1991 the numbers on government schemes were recorded separately but similar
figures were not produced for 1981 or 2001. This reflect the smaller number on such
schemes in 1981 and the fact in 2001 a higher proportion of government-supported
trainees were with employers rather on free-standing schemes. For 1991, men on
government schemes (27,300 in all across the English and Welsh coalfields) have
been added to the number of unemployed in their area of residence.
The 2004 figures are the number of claimant unemployed resident in coalfield wards
in April 2004, grossed up by the ratio between Census and claimant unemployment
by sex in 2001.
‘RESIDUAL’
Calculating each component of the accounts separately generates a residual in the
accounts as a whole, arising mainly from minor inconsistencies between different
datasets. The residual is small – equivalent to well under 0.5 per cent of the working
age population between 1991 and 2001 for example. For presentational purposes
the residual is incorporated into the net commuting figures, which tend in any case to
be subject to a wider margin of error than the other variables in the accounts
because of the larger number of individual elements that feed into the commuting
calculation.
In previous labour market accounts for the coalfields for 1981-91, presented in Beatty
and Fothergill (1996), the residual was incorporated into the change in economic
activity. The figures presented in this paper therefore differ a little from those
published previously. The change makes no important difference to the findings.
42
Incapacity benefit data
The incapacity benefit data used in this paper comes from the Department for Work
and Pensions (DWP). The data refers to men (and women) of working age who are
out-of-work and in receipt of Incapacity Benefit, NI credits for incapacity, or Severe
Disablement Allowance. Together, these comprise the non-employed group
conventionally referred to by government as receiving ‘incapacity benefits’. DWP
produces estimates by ward based on a 5 per cent sample of all benefit claimants.
Given the very large numbers claiming incapacity benefits, the ward-based estimates
are likely to be tolerably reliable, especially when aggregated to larger units such as
the coalfields.
Acknowledgements
This paper is the result of independent academic research. It does however draw on
the findings of the authors’ previous research funded by the Economic and Social
Research Council (into coalfield labour markets and into male economic inactivity),
by the East Midlands Observatory (on hidden unemployment) and more recently by
English Partnerships (to up-date coalfield boundaries and data for England). The
paper uses output from the Census of Population which is Crown Copyright and is
reproduced with the permission of the Controller of HMSO.
